Project description:Background: Wnt signaling maintains the undifferentiated state of intestinal crypt progenitor cells by inducing the formation of nuclear TCF4/beta-catenin complexes. In colorectal cancer, activating mutations in Wnt pathway components cause inappropriate activation of TCF4/beta-catenin-driven transcription. Despite the passage of a decade after the discovery of TCF4 and beta-catenin as the molecular effectors of the Wnt signal, few transcriptional activators essential and unique to the regulation of this transcription program have been found. Methodology/Principal Findings: Using proteomics, we identified the leukemia-associated Mllt10/Af10 and the methyltransferase Dot1l, as Tcf4/beta-catenin interactors in mouse small intestinal crypts. Mllt10/Af10-Dot1l, essential for transcription elongation, are recruited to Wnt target genes in a beta-catenin -dependent manner, resulting in H3K79 methylation over their coding regions in vivo in proliferative crypts of mouse small intestine, in colorectal cancer and Wnt-inducible HEK293T cells. Depletion of MLLT10/AF10 in colorectal cancer and Wnt-inducible HEK293T cells followed by expression array analysis identifies MLLT10/AF10 and DOT1L as essential activators dedicated to Wnt target gene regulation. In contrast, previously published b-catenin coactivators p300 and beta-catenin displayed a more pleiotropic target gene expression profile controlling Wnt and other pathways. tcf4, mllt10/af10 and dot1l are co-expressed in Wnt-driven tissues in zebrafish and essential for Wnt-reporter activity. Intestinal differentiation defects in apc-mutant zebrafish can be rescued by depletion of Mllt10 and Dot1l, establishing these genes as activators downstream of Apc in Wnt target gene activation in vivo. Morpholino-depletion of mllt10/af10-dot1l in zebrafish results in defects in intestinal homeostasis and a significant reduction in the in vivo expression of direct Wnt target genes and in the number of proliferative intestinal epithelial cells. Conclusions/Significance: We conclude that Mllt10/Af10-Dot1l are essential, dedicated activators of Wnt-dependent transcription, critical for maintenance of intestinal proliferation and homeostasis. The methyltransferase Dot1l may present an attractive candidate for drug targeting in colorectal cancer.
Project description:Background: Wnt signaling maintains the undifferentiated state of intestinal crypt progenitor cells by inducing the formation of nuclear TCF4/beta-catenin complexes. In colorectal cancer, activating mutations in Wnt pathway components cause inappropriate activation of TCF4/beta-catenin -driven transcription. Despite the passage of a decade after the discovery of TCF4 and beta-catenin as the molecular effectors of the Wnt signal, few transcriptional activators essential and unique to the regulation of this transcription program have been found. Methodology/Principal Findings: Using proteomics, we identified the leukemia-associated Mllt10/Af10 and the methyltransferase Dot1l, as Tcf4/beta-catenin interactors in mouse small intestinal crypts. Mllt10/Af10-Dot1l, essential for transcription elongation, are recruited to Wnt target genes in a beta-catenin -dependent manner, resulting in H3K79 methylation over their coding regions in vivo in proliferative crypts of mouse small intestine, in colorectal cancer and Wnt-inducible HEK293T cells. Depletion of MLLT10/AF10 in colorectal cancer and Wnt-inducible HEK293T cells followed by expression array analysis identifies MLLT10/AF10 and DOT1L as essential activators dedicated to Wnt target gene regulation. In contrast, previously published b-catenin coactivators p300 and beta-catenin displayed a more pleiotropic target gene expression profile controlling Wnt and other pathways. tcf4, mllt10/af10 and dot1l are co-expressed in Wnt-driven tissues in zebrafish and essential for Wnt-reporter activity. Intestinal differentiation defects in apc-mutant zebrafish can be rescued by depletion of Mllt10 and Dot1l, establishing these genes as activators downstream of Apc in Wnt target gene activation in vivo. Morpholino-depletion of mllt10/af10-dot1l in zebrafish results in defects in intestinal homeostasis and a significant reduction in the in vivo expression of direct Wnt target genes and in the number of proliferative intestinal epithelial cells. Conclusions/Significance: We conclude that Mllt10/Af10-Dot1l are essential, dedicated activators of Wnt-dependent transcription, critical for maintenance of intestinal proliferation and homeostasis. The methyltransferase Dot1l may present an attractive candidate for drug targeting in colorectal cancer. 6 samples for Ls174T cells: si-b-catenin against si-control and dyeswap of it, si-control, si-MLLT10, si-BRG1 and si-P300 are hybridized against common reference RNA; 6 samples of HEK293T cells: Wnt3A or control medium (CM) induction for 9 hours, si-MLLT10, si-DOT1L, si-BRG1 and si-P300 upon 9 hour Wnt3A induction are all hybridized against common reference RNA
Project description:We report calling of super-enhancers in CALM-AF10 mouse AML cells where the fusion is either on (DMSO) or turned off (using Doxycycline).
Project description:AF10 is a cofactor of the H3K79 methyltransferase DOT1L. To uncover the role of H3K79me in reprogramming to induced pluripotent stem cells (iPSCs), we bred reprogrammable mice encoding doxycycline inducible Oct4-2A-Klf4-2A-IRES-Sox2-2A-c-Myc (OKSM) transgene with conditional flox (fl)-AF10 (Mllt10). We isolated mouse embryonic fibroblasts (MEFs), deleted AF10 with CRE-recombinase or treated cells with empty vector control, and initiated reprogramming in DOT1L chemical inhibitor SGC0946 or DMSO control. Gene expression was assessed using RNA-Seq and genome wide localization of H3K79me1 and H3K79me2 was determined by ChIP-Seq on day 4 of reprogramming.
Project description:AF10 is a cofactor of the H3K79 methyltransferase DOT1L. To uncover the role of H3K79me in reprogramming to induced pluripotent stem cells (iPSCs), we bred reprogrammable mice encoding doxycycline inducible Oct4-2A-Klf4-2A-IRES-Sox2-2A-c-Myc (OKSM) transgene with conditional flox (fl)-AF10 (Mllt10). We isolated mouse embryonic fibroblasts (MEFs), deleted AF10 with CRE-recombinase or treated cells with empty vector control, and initiated reprogramming in DOT1L chemical inhibitor SGC0946 or DMSO control. Gene expression was assessed using RNA-Seq and genome wide localization of H3K79me1, H3K79me2, and RNA Polymerase II (RNAPII) was determined by ChIP-Seq on day 4 of reprogramming.
Project description:We performed pooled CRISPR screens to determine the whether the CALM-AF10 target genes and protein interactors identified by RNA-Seq and proteomics methods are functional dependencies for CALM-AF10-driven AML. We constructed a pooled CRISPR library of CALM-AF10 mouse target genes and interactors, and used it to transduce CALM-AF10 mouse AMLs. The cells were sampled at "Day 0" and kept in culture to determine in vitro dependencies at day 12. For in vivo screens, the input cell pool from day 0 was injected in mice and mice where sacrificed upon signs of AML disease, and the AML cells were harvested.
Project description:Interest focuses on genes encoding histone demethylases in hematologic malignancies, such as EZH2 (enhancer of zeste homolog 2). EZH2 mutations were recurrently observed in lymphomas and chronic myeloid malignancies, but data in acute leukemias are limited. We investigated 13 PICALM-MLLT10 (=CALM-AF10) rearranged acute leukemia predominantly of T-lineage (7 m/6 f; 6–53 years) by deep-sequencing for EZH2mut and identified 3 (23%) EZH2mut carriers: one splice site mutation in exon 14, while two patients had missense mutations in the D1 region of exon 5 which interacts with different DNA methyltransferase genes (but no DNMT3Amut was detected in the 13 PICALM-MLLT10-positive patients). In contrast, no EZH2mut was found in an independent cohort of 12 PICALM-MLLT10-negative T-ALL. Gene expression profiling revealed increased expression of genes with a role for transcription or intracellular transport processes in the PICALM-MLLT10-positive cases. The frequent occurrence of EZH2mut in PICALM-MLLT10-positive malignancies emphasizes a cooperative effect in acute leukemias.
Project description:Interest focuses on genes encoding histone demethylases in hematologic malignancies, such as EZH2 (enhancer of zeste homolog 2). EZH2 mutations were recurrently observed in lymphomas and chronic myeloid malignancies, but data in acute leukemias are limited. We investigated 13 PICALM-MLLT10 (=CALM-AF10) rearranged acute leukemia predominantly of T-lineage (7 m/6 f; 6–53 years) by deep-sequencing for EZH2mut and identified 3 (23%) EZH2mut carriers: one splice site mutation in exon 14, while two patients had missense mutations in the D1 region of exon 5 which interacts with different DNA methyltransferase genes (but no DNMT3Amut was detected in the 13 PICALM-MLLT10-positive patients). In contrast, no EZH2mut was found in an independent cohort of 12 PICALM-MLLT10-negative T-ALL. Gene expression profiling revealed increased expression of genes with a role for transcription or intracellular transport processes in the PICALM-MLLT10-positive cases. The frequent occurrence of EZH2mut in PICALM-MLLT10-positive malignancies emphasizes a cooperative effect in acute leukemias. 29 patients analyzed with gene expression microarrays.
Project description:Epigenetic regulators have important roles during embryonic development as well as somatic cell reprogramming. We previously showed that inhibition of DOT1L, the histone H3 lysine 79 methyltransferase, increases the efficiency of reprogramming via regulation of lineage specific genes. However, the role of DOT1L-interacting proteins in reprogramming remains unknown. In this study, DOT1L interactors were identified using the BioID method in which a promiscuous BirA ligase (BirA*) was employed to biotinylate DOT1L-proximal proteins. The resulting interaction candidates were investigated for their effects on reprogramming. Candidate genes were knocked-down in human fibroblasts via shRNAs followed by reprogramming. Our results indicated that knock-down of AF10 (MLLT10), significantly increased the iPSC generation efficiency, suggesting that it acts as a barrier to reprogramming similar to DOT1L. This finding was verified by CRISPR/Cas9 mediated knockout of AF10. Overexpression of AF10 reversed the effect of AF10 knockout and decreased reprogramming efficiency. To determine how AF10 silencing changes the gene expression, RNA-sequencing was performed on human fibroblasts undergoing reprogramming. AF10 suppression resulted in downregulation of fibroblast-specific genes and accelerated the activation of pluripotency-related genes. Our analysis also demonstrated that silencing of AF10 results in gene expression changes similar to DOT1L inhibition during reprogramming. Taken together, this study uncovered AF10 as a novel barrier to reprogramming and contributed to our understanding of epigenetic mechanisms that maintain cell identity.